Characterization Of The 2D/3D Feature Of Urban Buildings And Its Differences Of Seasonal Thermal Environmental Effects

The urban heat island effect,the phenomenon that the temperature in urban areas is higher than that in the surrounding countryside,has many adverse effects on the natural and social environments.The urban spatial structure is an important factor affecting the urban thermal environment,and urban buildings play an important leading role in the urban thermal environment by changing the urban energy exchange.Therefore,it is important to investigate the influence mechanism of complex urban characteristics on the distribution of spatial and temporal heterogeneity of urban thermal environment to mitigate urban heat island effect and promote sustainable urban development.In this paper,we select Changchun city within the fourth ring traffic boundary as the study area,use Landsat8 TIRS as the main thermal infrared remote sensing data source,and apply atmospheric correction algorithm to invert the surface temperature of2017,2018 and 2019 corresponding to four different seasons to characterize the surface thermal environment;use the 3D building vector data and the domestic high-definition second image as the data source to construct a 2D/3D building database at different scales.Using a combination of qualitative and quantitative methods,GIS spatial analysis,landscape pattern analysis,and classical statistical analysis were used to explore the differences of urban building 2D/3D spatial structure on seasonal thermal environment under multi-scale space.The main findings of this study are as follows.(1)From the spatial pattern of building characteristics: for two-dimensional indicators,the building coverage is mainly less than 0.28,and decreases with the increase of spatial scale;the areas with high density of building patches are mainly located in the central city,and the differences in the density of patches at different spatial scales are large,and decrease with the increase of spatial scale;the density of building perimeter shows a spatial distribution pattern of "dense in the south and sparse in the north".The building perimeter density shows a spatial distribution pattern of "dense in the south and sparse in the north",which increases with the increase of spatial scale;the building perimeter area ratio is disperse in the grid of 240 m and 480 m,mainly with a low perimeter area ratio less than 0.57.For the three-dimensional indicators,there are a total of 35575 buildings in the study area,among which medium and low-rise buildings less than 16.8m in height dominate,accounting for about 70%,and the areas with high building volume are mainly concentrated in the central part of Changchun,indicating that the inner city areas occupy more three-dimensional space volume,and the volume ratio is mainly lower than 2.0 in the low level,and the higher areas are concentrated in the central city area and the southwest industrial areas.(2)From the spatial pattern of surface temperature: there is a good consistency between the high temperature surface temperature and the spatial distribution of buildings in the study area.The high-temperature areas are mainly distributed in the automobile city in the southwest,the chemical plants in the southeast and the traffic hubs in the central and western parts of the city,while the low-temperature areas are mainly distributed near the Itong River and the South Lake Park in the south and the new urban areas in the southeast.Seasonally,the surface temperature difference is the largest in summer(26.1℃),and the area of heat island effect area(Urban Heat Island Rejion,UHIR)is the highest(31.43%);the surface temperature difference is 20.4℃ and15.3℃ in spring and autumn respectively,and the percentage of UHIR is 26.96% and27.68% respectively;the surface temperature difference is the smallest in winter(14.5°C)and the lowest UHIR percentage(26.59%)in winter.From the scale point of view:as the scale increases,the surface temperature in all seasons shows a trend of decreasing extreme difference,decreasing mean value,and decreasing data dispersion.(3)From the correlation between building characteristic indicators and thermal environment: two-dimensional indicators of building coverage have positive correlation with surface temperature in all four seasons and are the most important drivers of urban surface heat island effect,and building perimeter area ratio has negative correlation with surface temperature in summer,autumn and winter at the scale of 960 m.Threedimensional indicators of total building height and average building height showed significant negative correlations with surface temperature in all four seasons,and total building volume showed significant positive correlations with surface temperature in summer and autumn at 480 m and 960 m scales.Building shading has no significant effect on surface temperature in this study.(4)From the mechanism of the influence of building indicators on the thermal environment: the contribution of the two-dimensional building indicators to the surface temperature is greater than that of the three-dimensional building indicators in all seasons at all scales.For the two-dimensional indicators,the highest explanatory degree is found in the 480 m scale space(R~2 in four seasons: 0.344,0.362,0.385,0.192,respectively);the contribution of building coverage is the largest and most significant in autumn(standard coefficients in three scales: 0.876,0.900,0.794,respectively),and building perimeter density combined with other two-dimensional building indicators can have a negative impact on surface temperature.For the 3D indicators,the highest explanation is found at the scale of 960m(R~2 at four seasons: 0.362,0.297,0.359,0.213,respectively);the contribution of total building height is the highest,also in autumn(standard coefficients at three scales:-0.398,0.659,0.791,respectively);the volume ratio at the scale of 480 m combined with The volume ratio has a positive effect on the surface temperature at the 480 m scale in combination with other building 3D indicators.(5)This study shows that reducing the proportion of building coverage and perimeter density at three different spatial scales,and appropriately increasing the proportion of mid-rise buildings and reducing the volume ratio at the 960 m scale can mitigate the negative effects of urban heat island effect.The findings of this study help to deepen the knowledge and understanding of the impact of urban building spatial structure on surface temperature from a multi-temporal perspective,and provide a basis and ideas for urban planning,ecological environment improvement,and sustainable urban development.